The Norwegian Government Petroleum Fund may not bestride the world as a financial Colussus quite yet, but one day soon it will. At the end of its first year of operations in 1996, its assets were a respectable $6bn, three years later, their market value at year-end was around $30bn.
The Ministry of Finance, which is responsible for the fund, has projected that it could hit $43bn by the end of this year and $61bn by the end of 2001. But these figures do not fully reflect the massive rise in oil prices seen this year, which will dramatically increase the flows to the coffers of the Norges Bank – the country’s central bank – charged with the conversion of surplus oil revenues into non-domestic financial assets, which are managed by its investment arm Norges Bank Investment Management (NBIM).
The fund has a benchmark portfolio of 60% fixed income and 40% in an equity portfolio that, until last year, was all externally managed, again mainly on an indexed basis, apart from the futures trading. All of the assets are invested outside Norway for a number of reasons, including diversifying from the domestic economy and obtaining higher returns.
As Yngve Slyngstad, NBIM’s head of equities in Oslo, explains: “We get regular cash flow streams from the North Sea, which we put into three regional equity buffer funds. The first is benchmarked 50% FT-Europe, the second 30% FT-North America and the third 20% FT-Pacific.” But within the regions, there is a stipulated list of countries with the market cap weights being used. On the fixed income side, Salomon Smith Barney’s World Government Bond Index is used for 18 countries, with the regional split being the same as for equities.
On the equity side, the funding is done through a sophisticated transition account, which serves as the conduit to both the external and internal managers, at which stage only futures are bought. “We then move the funds move through the account, where the futures are sold and the physicals purchased,” explains Slyngstad. “These are then transferred to our different fund managers, who specify the stocks, which we transfer to them at the end of each month. So basically, we purchase their portfolios for them. The external fund managers specify the stocks that will be transferred, and NBIM purchases these while simultaneously selling down their futures exposure. The team has built its expertise in programme trades and crossings in order not to move the market.”
Another reason for wanting to control the funding operations is that the asset managers work on performance- related fees. “The ways of calculating returns in funding periods are imprecise, barring daily reconciliation, and can cause problems, particularly in volatile markets. In addition, there is not an agreed way of measuring execution costs, so funding benchmarks are difficult to provide,” he points out.
“What we have found, basically, is that we can do this better than most of the managers can, because we have enormous cash flows.” This has enabled them to build a desk of five people dedicated to trading. “They are sitting there managing the exposures the whole time – we now have the structures to handle these trades.”
Slyngstad admits that the art of handling single stock trades and finding the pools of liquidity, is not their forte. “The external managers are better at this than we are!” Consequently, the managers carry out trading in illiquid names themselves.
But another essential role for this team is to control the overall exposures of the fund, which has strict risk parameters laid down for it. “We have to deal with the market exposures by moving from physicals to futures or from physicals to physicals so that we can do transitions without affecting the total risks of the fund even for a few days. It is the same team who are controlling this and keeping the risk levels of the fund stable.”
The size of the futures trading undertaken by the fund easily exceeds $15bn a year, he says. “We have shown we have the capacity and capability to do this. But the reason for doing this internally is to maintain secrecy about what and when we are doing things.” The fund has the function of pushing assets into the market down to a fine art now. “This is now a continuous process, but if we want to do something on the big size, we want to keep our cards to ourselves.” But it is still a task of daunting maginitude. Revenues were originally projected to be $10bn this year, when oil was $16 a barrel, but Slyngstad says it will inevitably be much higher.
He adds that he does not think that funding external managers in kind, or central risk management through an overlay portfolio is something most plan sponsors necessarily should contemplate. NBIM developed in this way because of the funds’ needs and the size of the flows being handled.
The fund was given a clear mandate to achieve the ‘highest possible returns’ relative to the benchmark and the risk constraints imposed, one of the main ones being that the expected tracking error is not to exceed 1.5 percentage points. The balance between fixed income (50 to 70%) and equities (30 to 50%) can be varied in order to obtain outperformance and for rebalancing of the portfolios, says NBIM. Limits have also been set for interest rate risk and a 1% ceiling was originally imposed on the maximum holdings the fund can
have in any one company. This was increased to 3% last month
Slyngstad describes the approach that the fund has adopted as being a core satellite strategy, though he has some reservations: “What we have is this core of indexed funds and a top layer of active funds, covering the five geographical regions. These are all segregated accounts, all with a high risk profile with a 5% or more tracking error typically as a target.” Then, there is the internally managed sectoral specialist layer and this is likely to be developed further. Currently, it stands at about 10% of the equity holdings. Recently, NBIM awarded the first external active sector mandate and is looking for more sector specialists.” A third layer of private equity may be a future possibilty.
At the beginning of last year, around 90% of the equity portfolio was on an indexed basis, with Barclays Global Investors and Deutsche Asset Management running global indexed portfolios and Gartmore looking after a UK portfolio.
On the active side, Capital International, Gartmore and Storebrand were running Europe ex-UK portfolios, with Mercury handling a UK brief and Fidelity Pensions Management running Asia and Oceania, excluding Japan. During the year Capital International and Fidelity picked up active Japan mandates and mid year the in-house team came on stream as from June with a sectoral brief covering financial, IT and non-cyclical services. At year end, some 28% of the equity portfolio was being actively managed, with the prospects of this being increased to 40% this year.
“Technically, we like to look at it in terms of risk or tracking error. We are moving from a risk level of 50 basis points to 100 – basically doubling the risk during the year. The reason for this change in thinking is that we believe we can identify active managers who can deliver excess performance – it’s as simple as that.”
The fund has different risk approaches according to region with the majority of active risk being taken in continental Europe and Japan. So, at the end of 1999, while the portfolios were 100% indexed as far as north America was concerned, it was only 40% so in continental Europe. Slyngstad reckons about half of the portfolios will be actively managed, but he questions the impact that will have: “So even with 28% active active mandates in our equity portfolios, only half of that will deviate from the index as active bets.”
With such an array of managers and strict risk parameters, great importance is attached to the central risk management function. “We have an overlay portfolio, which takes on a daily basis data from the 28 segregated accounts and pulls all this together. If we find there is an excess return on a manager in Japan, we will be selling futures in Japan, or if a manager has cash we will equitise the cash. There is the question of index drifting since all the specialist mandates do not aggregate, you will have to buy or sell futures in the rebalancing fund.”
The data comes from the managers overnight through the fund’s global custodian Chase’s system and is down-loaded the next morning. This data is down to stock specific level as this information is needed in order to be able to comply with the 3% limit on shareholdings. “This issue makes it essential to have the information on a daily basis. We use the overlay portfolio to balance out any risks we do not want – we are happy to have risk, but not those we did not want to buy.”
The balance between internal and external is purely a question of performance, he says. “Ideally, we will fund index, active external and active internal from the point of view of the quarterly inflows, depending on who has done best in the last four rolling quarters. If you cannot develop a fund management capability internally that is competitive, then do not do it.” The monitoring that the internal managers are exposed to is exactly the same as for the external. “As to how the sector versus geographical or regional split develops, we are completely pragmatic, it is just a matter of where we see the possibilities of creating excess performance.”
The fund’s key activities will for control purposes always be based in Oslo, such as the overlay programme. But the fund’s operations are split location-wise, with an office in New York, where much of the fixed income management is handled, a back office in Edinburgh and now a new base in London.
The selection of external manager is an ongoing process for the fund, which has recently appointed fixed interest managers after assessing proposals from 80 hopefuls. Similarly, tactical asset allocation managers have been appointed earlier this year, after 20 managers sought the mandate.
The selecting and monitoring of both active and indexed managers is something the fund intends to master.
“What you are trying to do is assess whether a particular manager can beat the market.” But that depends on what market is being discussed, as it is quite a different matter to outperform in the UK compared with Japan.
The initial screening is across 10 criteria, the most important of which, he says, is the average realised information ratio. “We look at how their portfolios have changed on a monthly basis for five years. All the responses to the questionnaires are put into a huge data base and we analyse information ratios and portfolio construction.” The tricky bit, he acknowledges, is to estimate the expected excess return. “This is difficult because you cannot just say you expect this manager to outperform by 1%. So we try to take into account a number of different aspects here, such as the portfolio trading costs, expected brokerage costs, expected turnover, expected fees, assets in the regions and the assets in these products.” A key factor is that of moving the market, because if managers are making the same buy for a number of portfolios, it increases the chances of moving the markets, he points out. “We also have a ‘k’ factor, as to how well they are able to find the pools of liquidity”.
One important issue for the fund is that of correlation between active managers. “Does it make a difference if we have two, three or four? What we have generally found is that it does not pay to have too many active managers. But it is essential to figure out their transaction costs, and the full market impact costs of their trading. We are very focussed on these aspects. This is an area where active managers can improve as they have often used up their advantages in costly implementation. The index manager has definitely got the upper hand here.”
The question then moves on to monitoring the managers, once they have been selected and put together to, hopefully, optimal combinations or baskets. The system is being designed to monitor the internal and external managers in the same way. “This is still a learning process for us as we are only two years into equity investment and we feel we have a long way to go. We now have 12 monitoring modules for looking at the external portfolios.”
In addition, the fund regularly goes through all managers’ portfolios on a weekly basis: “ This is to see if there are any reasons for what the managers are doing, seeing who is ahead of the brokers, who is moving after them, who is moving ahead or after the other managers and to see if they are just market followers or contrarians.”
The aim is to get the same feel for the portfolio as the portfolio manager in the external organisation. “We want to know as much about our portfolio, as the portfolio manager does.”
In 1999, the fund had an overall return of 12.4%, with 35.8% coming from the equity portion and -0.99% from the fixed income. The equity index managers returned 32.57% just below benchmark by 0.03%, while the active managers in Europe and the UK at 35.96% exceeded benchmark by 10.32 percentage points. At the same time, the active Asian and Oceanic portfolios at 81.93% produced excess returns of 29.24% in the nine months of running. The overall excess return over benchmark was 1.1% for the total fund, which the fund attributes to the good results from the external active equity managers, and comments that it was markedly better than expected. It reckons its cost in 1999 came to 0.09% of assets, which it contrasts to the 0.10% charge made the giant pension funds ABP in the Netherlands and CalPERS in the US.
“We know what we are doing. It is not perfect, but we want to make it as perfect as we can,” says Slyngstad.
If Norway’s Petroleum Fund can’t make it perfect, who can?